Consumption of alcoholic beverages in Brazil: estimation of prevalence ratios – 2013 and 2019

ABSTRACT OBJECTIVES To estimate the prevalence of weekly, monthly and abusive alcohol consumption in Brazil in 2013 and 2019, compare the period estimates, and verify the magnitude of the differences. METHODS Analysis of data on alcohol consumption in the adult population (18 years or older) from the National Health Survey (PNS), 2013 and 2019. The number of interviewees in 2013 was 60,202 and 88,531 in 2019. The samples were characterized according to demographic, socioeconomic, health, and alcohol consumption variables and differences in proportions in the period were compared using Pearson’s c2 test, with Rao-Scott approximation and a 5% significance level. Multivariate Poisson regression models were estimated for the outcome variables of monthly, weekly and abusive consumption of alcoholic beverages, in order to estimate the magnitude of the differences between the 2013 and 2019 PNS estimates, using the prevalence ratio (PR). Models were adjusted per sex and age group and stratified per sex and demographic region. RESULTS There was a difference in the distribution of the population according to race, occupation, income, age group, marital status, and education. There was an increase in alcohol consumption for all outcome variables, with the exception of weekly consumption in males. The PR of weekly consumption was 1.02 (95%CI 1.014–1.026), and in females the PR was 1.05 (95%CI 1.04–1.06). The highest PRs in the general population and per sex occur for abusive consumption. The increase in weekly consumption per region occurred in the South, Southeast, and Central-West regions. CONCLUSIONS Males are the main alcohol consumers in Brazil; the PRs for both males and females show that there was an increase in monthly, weekly and abusive consumption in the research period; it is noteworthy that females have increased their consumption pattern with greater intensity than males.

摘要

研究目的：估算2013年与2019年巴西成年人群周饮酒、月饮酒及有害饮酒的流行率，对比两时期的估算结果，并验证差异的幅度。

研究方法：本研究分析2013年与2019年全国健康调查（National Health Survey, PNS）中≥18岁成年人群的饮酒相关数据。2013年与2019年的受访人数分别为60202人与88531人。研究根据人口学、社会经济学、健康状况及饮酒相关变量对样本进行特征描述，并采用拉奥-斯科特（Rao-Scott）近似的皮尔逊卡方（Pearson’s χ²）检验，以5%的显著性水平对比两时期的构成比差异。以患病率比（prevalence ratio, PR）为效应指标，针对月饮酒、周饮酒及有害饮酒这三类结局变量构建多变量泊松回归模型，以此估算2013年与2019年PNS数据中各结局的差异幅度。模型按性别与年龄组进行调整，并按性别及人口学区域进行分层分析。

研究结果：不同种族、职业、收入、年龄组、婚姻状况及教育水平人群的分布存在差异。除男性周饮酒率外，其余各类饮酒结局的发生率均有所上升。男性周饮酒的PR值为1.02（95%CI：1.014~1.026），女性则为1.05（95%CI：1.04~1.06）。在全人群及分性别分析中，有害饮酒的PR值均为最高。周饮酒率上升的区域为南部、东南部及中西部地区。

结论：巴西人群中男性为主要饮酒群体；男女各类饮酒结局的PR值均显示，研究周期内月饮酒、周饮酒及有害饮酒的发生率均有所上升；值得注意的是，女性饮酒模式的增幅强度高于男性。